The history of TCP/IP is somewhat muddy, as you can imagine. At 02:04 PM 3/27/02, Steven A. Ridder wrote: >I am a technical reviewer for a book, and someone wrote that TCP/IP was >written by the Depertment of Defense.
I agree that you should question that. > I am confident that ARPAnet was >commissiond by the DoD in the 60's to BBN Yes, you could say that. The Information Processing Techniques Office (IPTO) of ARPA awarded the contract to build the Interface Message Processors (IMP) for ARPANET to BBN in late 1968. IMPs were the early routers. BBN built the IMPs with the help (or hindrance if you believe some reports) of Honeywell. Honeywell developed and manufactured the hardware. BBN did the software. Descriptions of the "network layer" software that ran on these IMPs doesn't sound much like IP at all. It was connection-oriented, for one thing, and handled error correction. It was very East-Coast anal-retentive stuff. ;-) The software that evolved into TCP/IP was a West-Coast hippy-dippy geeky phenomenon. UCLA, SRI, UC Santa Barbara, USC, and University of Utah graduate students and researchers worked on it. Originally they had to make sure their software interoperated with the IMPs of the ARPANET. They developed a protocol called the Network Control Protocol (NCP) that worked on the end devices that communicated with the IMPs. It was a host-to-host protocol that could be considered a predecessor to TCP. NCP worked only with ARPANET. By 1973 or so, ARPANET wasn't the only game in town though. There was packet radio (which evolved into Ethernet), SATNET, and others. A more general-purpose protocol was needed. Vint Cerf who was with UCLA at the time and Bob Kahn, who had been at BBN but now worked for ARPA directly, worked on a new protocol called Transmission Control Protocol (TCP) that was general-purpose. They made the assumption that the underlying network was unreliable. The new protocol shifted the job of reliability from the network to the destination hosts. Originally TCP handled the routing of packets also. TCP had jobs that we would today assign to the network and transport layers. And finally, in 1978, we come to the birth of the Internet Protocol (IP). In 1978, the job of routing packets was broken away from TCP. TCP was given the task of breaking messages into packets, reassembling them at the other end, detecting errors, resending anything lost, and putting packets in the right order. IP was simply responsible for forwarding individual packets. The specifications for how this should work were written by Cerf at UCLA, and Postel and Cohen from the University of Southern California's Information Sciences Institute (ISI). In the early 1980s, the ARPANET got really congested and the National Science Foundation created its own network for the academic computer science community. It used TCP/IP and is sometimes considered the real forerunner of "the Internet," although it probably could never have happened without the work that went into the ARPANET. ARPANET converted to TCP/IP in 1983. It also divided into MILNET and ARPANET. It had connectivity with all the other networks by then. Later it got decommissioned. By 1989, it was gone, but its legacy lived on. May it RIP. ;-) Here's a recommendation for a terrific book about the history of the Internet: "Where Wizards Stay Up Late: The Origins of the Internet" by Katie Hafner and Matthew Lyon. Priscilla >, and maybe TCP/IP was derived from >these early protocls, but to say the the DoD, or BBN or anyone other than >the Internet community wrote TCP and IP would be incorrect, right? I seem >to remember that IP was used in ArpaNet, but not TCP. I thought TCP was >written in various universities. I could even look up the couple (who used >to work at Cisco) who wrote it. > >-- > >RFC 1149 Compliant. >Get in my head: >http://sar.dynu.com ________________________ Priscilla Oppenheimer http://www.priscilla.com Message Posted at: http://www.groupstudy.com/form/read.php?f=7&i=39681&t=39657 -------------------------------------------------- FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
